Multi range Real-time depth inference from a monocular stabilized footage using a Fully Convolutional Neural Network.
Clement PinardLaure ChevalleyAntoine ManzaneraDavid FilliatPublished in: CoRR (2018)
Keyphrases
- real time
- convolutional neural network
- depth cues
- wide range
- low cost
- vision system
- depth cameras
- visual odometry
- pose estimation
- machine learning
- visual slam
- face detection
- depth map
- image sequences
- probabilistic inference
- bayesian networks
- depth information
- neural network
- model based tracking
- disparity range
- monocular images
- depth images
- stereo camera
- bayesian inference
- range data
- stereo matching
- multi view
- high resolution
- optical flow
- object recognition